December 15, 1998
Viral load tests allow physicians to track with greater accuracy than ever before the progression of HIV in the body -- thus helping their HIV-infected patients make choices about appropriate treatment strategies. The viral load test is only appropriate for a few specific situations, described below. Most people concerned about HIV do not need viral load testing. The antibody test is still the cheapest, easiest, and overall most reliable way for individuals to learn their HIV status.
There are two kinds of viral load tests. They both measure the amount of HIV RNA in a sample. (RNA is the "blueprint" that HIV uses to make more virus.) When comparing viral load test results to determine a trend, the same type of test must be used each time. It is otherwise like trying to compare the proverbial "apples" with "oranges."
One test, called the branched-chain DNA test (bDNA), is made by a company called Chiron (pronounced "KAI-ron"). The reverse transcriptase polymerase chain reaction or RT-PCR (commonly called the PCR) test is made by Roche (pronounced "RO-sh").
Scientists know how HIV RNA is constructed; it has a distinctive "pattern." By creating a mirror image of that structure and matching it against what they find in a blood sample, they can measure HIV RNA.
Why is indirect or direct measurement important? Because the same sample, tested with both tests, will give different viral load measurements. Here's another way to think of the differences between these two tests: One way to describe weather is to state the temperature. A thermometer is a common gauge for measuring temperature. It can measure temperature in degrees Celsius or in degrees Fahrenheit. Both measurements are "accurate," but they describe the same weather differently (when it is 25 degrees Celsius, it is 72 degrees Fahrenheit).
The same is true for HIV viral load testing. Both bDNA and RT-PCR tests measure the amount (or "load") of HIV in the blood. The test methods ("thermometers") are different, but what they are measuring (the "weather") is the same.
Viral load varies, sometimes a lot, over time. One viral load test, interpreted by itself, is not meaningful. It must be looked at with other tests, such as T-cell (CD4) count, or compared to other viral load tests to be helpful as an indicator of HIV progression.
The results of viral load tests are usually given as "copies per milliliter (ml)" of blood, like the CD4 (T-cell) count. Each virus carries two copies of RNA. If there are 100,000 copies of HIV RNA, that means 50,000 virus particles (or virions) are present. Currently:
Test results can be variable (different), even when repeated on the same blood sample. Any test report should also indicate the variability of the test. For example, the result may be 20,000 copies with a variability of 5,000. The viral load is then somewhere between 15,000 and 25,000 copies.
There is a lot of HIV in other places in the body, not just the blood. Only a fraction of HIV is in circulating blood; the rest is in the lymph system and other body tissues. Other factors may influence the variability of a test; viral infections, such as a cold or the flu, can cause a temporary increase in viral load. For this and other reasons, physicians usually request two or more tests over a short period of time (within two to four weeks) to establish baseline viral load.
As with the ELISA HIV antibody test and other medical tests, "accuracy" is a combination of sensitivity and specificity. Sensitivity of a test is its ability to detect the virus when it is present, and to not report "false negative." Specificity of a test is its ability to detect only HIV and nothing else, so as not to report false positive. Ideally, a test is both very sensitive and very specific. In practice, increasing sensitivity decreases the specificity, and vice versa.
Viral load testing is extraordinarily sensitive, but it is not perfect. The most sensitive viral tests can only detect 40 or more viral particles per milliliter in a sample. It is therefore false to assume that an "undetectable viral load" means there is "no HIV present." A person with "undetectable viral load" can indeed transmit the virus to someone else.
The downside to PCR's sensitivity is its somewhat lower ability to eliminate false positive results caused by RNA other than HIV. This incredible sensitivity also means that the smallest inattention while cleaning lab equipment after a previous infected sample could contaminate a negative sample and cause a false positive.
Physicians presently use baseline viral load to help their HIV-infected patients make choices about appropriate treatment strategies. Patients and their doctors can assess the need to adjust treatment based on their ongoing observations and monitoring of changing viral load levels over time. Because viral load testing is still a relatively new diagnostic tool, baseline viral load is used as one indicator of viral activity, but is not yet determined to be an absolute. Generally, most practitioners consider "low" baseline viral load to be approximately 500 or fewer copies of HIV RNA, and "high" to be any number above 40,000 copies. It is further believed that viral load may be a good marker or predictor of disease progression.
There is a correlation between CD4 level and viral load. Most people with higher CD4 cell counts have lower viral loads; conversely, people with higher viral load tend to have lower CD4 cell counts. (CD4 level is one indicator of how well the immune system is functioning.) Monitoring both CD4 level and viral load provides a more complete picture of immune health than using either test alone.
The viral load test is only appropriate for a few specific situations. Most people concerned about HIV do not need viral load testing. The antibody test is still the cheapest, easiest, and overall most reliable way for individuals to learn their HIV status.
Viral load testing is appropriate in the following circumstances:
Note: In each case but the first, the person will likely undergo a qualitative PCR test, which is related to a RT-PCR. The qualitative PCR does not measure the amount of HIV RNA, it detects the mere presence of HIV. It is so sensitive that it can detect one virus particle in a sample of 100,000 cells.
So, if it is so sensitive, why isn't PCR used for HIV diagnosis for everyone? Read the section on inappropriate use of viral load testing below.
Any physician can order a viral load test. Contact your own physician, or a community clinic.
The test can cost around $150 if you have to pay for it out-of-pocket. However, HIV-positive people (or others with valid reasons for needing the test, discussed above) who have Medi-Cal (in California) may be able to get it free through their local public health department or county hospital. Various research projects may offer free PCR testing if an individual is part of that specific study.
The vast majority of people concerned about HIV infection do not need viral load testing. In nearly all cases, this is not a test we recommend for assessing one's HIV status.
In many cases, people who want PCR testing to determine their HIV status are highly concerned/anxious. They are unwilling to wait through a three to six month window period and take an antibody test. They may have heard that PCR testing will indicate HIV status sooner than antibody testing. In some clinical research settings, this is true. Why only in research settings?
In research settings, other "controls," such as multiple tests and studying only those people who are confirmed to have been at high risk, reduce (but do not eliminate) the chance of making an error. However, HIV antibody testing, performed after the window period, is still required to confirm PCR test results, even in those highly controlled research environments.
Other reasons to not recommend viral load/PCR testing for HIV diagnosis: